Exploring the Potential of NoSQL Databases for a BIM and 3D GIS enabled Location Framework for Construction Digital Twins and the Golden Thread

Abstract. As digital transformation accelerates in the construction industry, combining data from Building Information Modelling (BIM) and 3D Geographic Information Systems (3DGIS) has become critical for enhanced decision-making across sectors including construction planning, supply chain management, health and safety and sustainable waste disposal. In combination, BIM and GIS underpin a location framework for Digital Twins across these applications. While traditionally relational databases have been used to integrate the data into one system, they face increasing limitations in handling the massive, complex BIM and 3DGIS data. This research investigates the potential of NoSQL databases as an alternative solution. Focusing on MongoDB, the study conducts a systematic performance comparison with the mature relational database, PostgreSQL. BIM and 3DGIS datasets from a live infrastructure construction project are utilised in three experiments assessing direct query, API-based retrieval, and 3D visualisation. The results provide valuable insights into MongoDB’s strengths in managing large spatial data via flexible schemas. However, limitations surface as data complexity and volume increases. Overall, this study concludes that NoSQL databases do offer some advantages for optimising BIM-3DGIS integration for seamless, combined, use, but more work is needed to fully harness their capabilities.